Update app.py

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 import numpy as np
 import pandas as pd
 import torch
+import time
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
 # ---------- Model ----------
@@ -17,6 +18,8 @@ model = AutoModelForCausalLM.from_pretrained(
 # ---------- Trend Logic ----------
 def detect_trend(values):
     diffs = np.diff(values)
+    if len(diffs) == 0:
+        return "FLAT"
     if all(d > 0 for d in diffs):
         return "INCREASING"
     elif all(d < 0 for d in diffs):
@@ -24,6 +27,7 @@ def detect_trend(values):
     else:
         return "MIXED"
 
+
 def change_score(values):
     values = np.array(values, dtype=float)
     if len(values) < 2:
@@ -34,15 +38,18 @@ def change_score(values):
 
 # ---------- LLM Explanation ----------
 def explain(kpi, values, trend):
+    # ✅ limit values to max 50 numbers
+    short_values = values[:50]
+
     prompt = f"""
-You are a telecom KPI expert.
-KPI name: {kpi}
-Values: {values}
-Detected trend: {trend}
-
-In 2 short sentences explain what this trend might mean.
-"""
-    inputs = tokenizer(prompt, return_tensors="pt")
+    You are a telecom KPI expert.
+    KPI name: {kpi}
+    Values (sample): {short_values}
+    Detected trend: {trend}
+    
+    In 2 short sentences explain what this trend might mean.
+    """
+    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=2048)
     output = model.generate(**inputs, max_new_tokens=120)
     return tokenizer.decode(output[0], skip_special_tokens=True)
 
@@ -57,30 +64,51 @@ def load_csv(file):
     return df, gr.update(choices=numeric_cols, value=numeric_cols[:5])
 
 # ---------- Main Analysis ----------
-def analyze(df, selected_kpis):
+def analyze(df, selected_kpis, progress=gr.Progress(track_tqdm=False)):
+    start_time = time.time()
     print("✅ Analyze function triggered")
 
     if df is None or not selected_kpis:
         return "❌ Upload CSV and select KPI columns"
 
+    progress(0, desc="Starting analysis...")
+
     results = []
     explanations = []
+    total = len(selected_kpis)
+
+    # ---- KPI Analysis ----
+    for i, kpi in enumerate(selected_kpis):
+        progress((i + 1) / total, desc=f"Analyzing {kpi}...")
 
-    for kpi in selected_kpis:
         vals = df[kpi].dropna().values.tolist()
         trend = detect_trend(vals)
         score = change_score(vals)
+
         results.append((kpi, trend, score, vals))
 
+    # ---- Sort by score ----
     ranked = sorted(results, key=lambda x: x[2], reverse=True)
 
-    # LLM explanation for top 5
+    # ---- LLM Explanations for Top 5 ----
+    progress(0.9, desc="Generating LLM explanations...")
+
     for kpi, trend, score, vals in ranked[:5]:
-        exp = explain(kpi, vals, trend)
+        try:
+            exp = explain(kpi, vals, trend)
+        except Exception as e:
+            exp = f"⚠️ LLM Error: {str(e)}"
         explanations.append(f"**{kpi}** → {exp}")
 
-    # Output
-    text = "## 🔥 Top 5 KPIs with Biggest Change\n"
+    progress(1.0, desc="Done ✅")
+
+    # ---- Time Taken ----
+    elapsed = round(time.time() - start_time, 2)
+
+    # ---- Output ----
+    text = f"⏱️ Time taken: {elapsed} seconds\n\n"
+    text += "## 🔥 Top 5 KPIs with Biggest Change\n"
+
     for kpi, trend, score, _ in ranked[:5]:
         text += f"**{kpi}** → {trend} (Score: {score})\n\n"
 
@@ -109,11 +137,11 @@ with gr.Blocks() as demo:
     output = gr.Markdown()
 
     analyze_btn.click(
-    fn=analyze,
-    inputs=[df_state, kpi_select],
-    outputs=output,
-    show_progress=True
+        fn=analyze,
+        inputs=[df_state, kpi_select],
+        outputs=output,
+        show_progress=True
     )
 
-    # analyze_btn.click(fn=analyze, inputs=[df_state, kpi_select], outputs=output)
+# ✅ Share link required in HF Spaces
 demo.launch(share=True)